A method for extracting and identifying surface morphology features of optical components

Abstract

The present invention discloses a method for extracting and identifying surface morphology features of an optical element, comprising the following steps: obtaining two-dimensional surface morphology information; identifying and filtering local morphology features; calculating a two-dimensional power spectrum density function, and obtaining a one-dimensional power spectrum density function by radial averaging; linearly fitting the one-dimensional power spectrum density function under double logarithmic coordinates to obtain a slope p and an intercept q; and calculating a two-dimensional extraction factor C of isotropic and anisotropic components of the surface morphology according to the fitting results. iso , C anis ; Inverse calculation of the two-dimensional Fourier transform components of the isotropic and anisotropic components of the surface morphology #imgabs0# and #imgabs1# Use the inverse Fourier transform to obtain the isotropic and anisotropic two-dimensional surface morphology components h recons Finally, the two-dimensional morphology components are evaluated. The present invention adopts the above-mentioned method for extracting and identifying the surface morphology features of an optical element, which can accurately extract and separate the coupled isotropic and anisotropic components, and can be used for parameter analysis and endpoint judgment of surface processing methods such as chemical mechanical polishing.

Description

Technical Field

[0001] The invention relates to the technical field of optical detection, and in particular to a method for extracting and identifying surface morphological features of optical elements. Background Art

[0002] After the surface of an optical component is processed, it is usually necessary to inspect and evaluate it. At present, a relatively complete inspection and evaluation system has been formed, such as a surface roughness evaluation system based on indicators such as the arithmetic mean, root mean square, and peak-to-valley value of surface height, and a surface texture and spatial feature evaluation system based on power spectral density (PSD) analysis and wavelet analysis.

[0003] However, the existing surface roughness evaluation system is mainly based on isotropic statistical methods, which has limited evaluation capabilities for anisotropic machined surfaces and tends to ignore directional features, resulting in inaccurate evaluation results. For example, in anisotropic machining methods such as milling and turning, the surface features have obvious directionality, and these roughness indicators often cannot fully reflect these directional features. Conventional PSD analysis methods also have limitations, which can cause many important structural information in the original data to be lost during the conversion process, especially for anisotropic and weakly periodic surface structures, the peaks in the PSD curve are often not obvious. This is because the loss of information in the vertical dimension in the PSD calculation makes these features difficult to identify, significantly affecting the accuracy of the analysis results. In addition, PSD analysis is very sensitive to the directionality of the surface structure, and changes in the rotation or detection angle will also affect the results, resulting in inconsistent and non-repeatable results. The decomposition results of wavelet analysis are also affected by the predetermined wavelet bands, which are prone to introduce artifacts, especially when dealing with complex surface morphologies. These artifacts may interfere with the identification of actual surface features and affect the accuracy of roughness evaluation. In order to overcome the limitations of the above surface evaluation system, it is urgent to propose a new surface feature separation and extraction method to achieve accurate identification of isotropic and anisotropic features.

[0004] In addition to the limitations of the above evaluation system, there may be local features in the original surface data, such as random scratches and holes. The coupling of these local features with the global morphology will reduce the texture contrast, affect the signal-to-noise ratio of the image, and increase the difficulty of distinguishing anisotropic feature signals. When evaluating images with low signal-to-noise ratios, if no noise filtering preprocessing is performed, the elimination of local features will rely on the experience and judgment of the inspectors, which will greatly reduce the efficiency and accuracy of product inspection in industrial production. Therefore, the extraction and filtering of local surface noise is crucial for the preprocessing of surface morphology. Summary of the invention

[0005] The purpose of the present invention is to provide a method for extracting and identifying the surface morphology features of optical elements. The method is universal for optical surface detection obtained by most processing methods, can accurately extract and separate coupled isotropic and anisotropic components, and directly evaluate the surface morphology features of different components.

[0006] To achieve the above object, the present invention provides a method for extracting and identifying surface morphology features of an optical element, comprising the following steps:

[0007] S1. Prepare surface samples with different morphological characteristics;

[0008] S2, obtaining two-dimensional surface topography height information through a surface topography detection instrument;

[0009] S3, local morphological feature recognition and filtering;

[0010] S4, applying the Welch window to reduce sampling errors and calculating the two-dimensional power spectral density PSD results;

[0011] S5. Use radial averaging method to obtain one-dimensional power spectral density PSD n result;

[0012] S6. One-dimensional power spectral density PSD under linear fitting double logarithmic coordinates n As a result, the slope p and intercept q of the fitting curve are obtained;

[0013] S7, according to the one-dimensional power spectral density PSD n Results, fitting slope p and intercept q, calculation of the two-dimensional extraction factor C of the isotropic and anisotropic components of the surface morphology iso , C anis ;

[0014] S8. Inverse calculation of the two-dimensional Fourier transform components of the isotropic and anisotropic components of the surface morphology and

[0015] S9, using inverse Fourier transform to obtain the isotropic and anisotropic two-dimensional surface morphology components h recons ;

[0016] S10. Evaluate the extracted isotropic and anisotropic two-dimensional morphological components.

[0017] Preferably, in step S1, the preparation of surface samples with different morphological features includes preparing an isotropic fractal surface by CMP, preparing a quasi-one-dimensional anisotropic surface by diamond turning, and preparing a two-dimensional anisotropic surface by ion beam polishing.

[0018] Preferably, in step S3, the local morphological feature recognition and filtering includes the following steps:

[0019] S31, use area bearing ratio to identify and filter random holes;

[0020] The calculation formula of the area bearing ratio is:

[0021]

[0022] Where F is the integral of the surface topography height distribution, f is the material area load ratio, and P(h≤h i ) represents the surface topography height h is less than h i The probability function of

[0023] S32, using Hough transform to transform the scratches into MC space, identify the peaks and remove random scratches, specifically:

[0024] The surface topography is transformed from the Cartesian coordinate system to the MC space, the scratch is modeled as a straight line y=mx+c, and the scratch is identified and removed by the peak value in the MC space.

[0025] Preferably, in step S4, the Welch window is applied to reduce the sampling error and calculate the two-dimensional power spectrum density PSD result, specifically: the Welch window is applied to reduce the blur and leakage caused by the limited sampling length and discrete sampling points, and the result of the calculated two-dimensional power spectrum density PSD is:

[0026]

[0027] Among them, the reconstruction normalization function w satisfies:

[0028]

[0029] in, represents the Welch window function, Represents the rectangular window function.

[0030] Preferably, in step S5, the one-dimensional power spectral density PSD is obtained by using the radial averaging method. n The results, specifically:

[0031] The radial average of the two-dimensional power spectral density PSD after noise filtering is as follows:

[0032]

[0033] Among them, PSD n is the two-dimensional PSD result after noise filtering, ρ L and ρ L-1 The coordinate point (u, v) satisfies in the polar coordinate system The polar diameter, NL Indicates the number of points within the region.

[0034] Preferably, in step S6, the one-dimensional power spectral density PSD under the linear fitting double logarithmic coordinates n As a result, the slope p and intercept q of the fitting curve are obtained, and the PSD after fitting is n The curve is expressed in the following form:

[0035] log(PSD n (ρ L ))=p·log(ρ L )+q.

[0036] Preferably, in step S7, the one-dimensional power spectral density PSD n Results, fitted slope and intercept, two-dimensional extraction factor C of the isotropic and anisotropic components iso , C anis writing:

[0037]

[0038]

[0039] Preferably, in step S8, the two-dimensional Fourier results of the isotropic and anisotropic components of the surface morphology are inversely calculated. and Specifically:

[0040] The following two-dimensional Fourier components are obtained by calculating the extraction factors in the Fourier domain:

[0041]

[0042]

[0043] Preferably, in step S9, the isotropic and anisotropic two-dimensional surface morphology components h are obtained by using inverse Fourier transform. recons , the specific calculation process is:

[0044]

[0045] in, is the reconstructed two-dimensional Fourier component, and IFT stands for inverse Fourier transform. Substituting the reconstructed two-dimensional Fourier component into the above formula, the reconstructed surface of the corresponding component is obtained.

[0046] Preferably, in step S10, the extracted isotropic and anisotropic two-dimensional morphological components are evaluated separately, specifically:

[0047] For the isotropic component, its fractal dimension and surface morphology roughness parameters are the main evaluation indicators; for the anisotropic component, the peak value, characteristic periodicity and directionality of the PSD curve are the main evaluation indicators.

[0048] Therefore, the present invention adopts the above-mentioned method for extracting and identifying the surface morphology features of an optical element, which has the following beneficial effects compared with the prior art:

[0049] (1) The surface pretreatment in the present invention can effectively filter random holes and random scratches on the surface, increase the contrast of effective information, improve the signal-to-noise ratio, and reduce the coupling degree of isotropic and anisotropic components.

[0050] (2) The present invention is capable of separating isotropic and anisotropic surface components, reconstructing reliable two-dimensional morphological information with improved contrast, and intuitively evaluating the morphological characteristics of each component.

[0051] (3) The present invention can directly obtain isotropic fractal information and anisotropic periodicity and directional parameters, thereby improving the efficiency and accuracy of surface detection and reducing the experience requirements for detection personnel.

[0052] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the overall process of an embodiment of a method for extracting and identifying surface morphology features of an optical element according to the present invention;

[0054] Figure 2 The present invention is a method for extracting and identifying the surface morphology features of an optical element according to an embodiment of the present invention. The results of the partial test of the profiler at 10× magnification at different polishing times are shown in FIG. 1 , wherein (a) is the partial test result of the profiler at 10× magnification at 0 min, (b) is the partial test result of the profiler at 10× magnification at 20 min, (c) is the partial test result of the profiler at 10× magnification at 60 min, (d) is the partial test result of the profiler at 100 min, and (e) is the partial test result of the profiler at 10× magnification at 180 min.

[0055] Figure 3 is the PSD curve of the surface in the X and Y directions of an embodiment of a method for extracting and identifying the surface morphology features of an optical element of the present invention, wherein (a) is the one-dimensional power spectral density PSD in the X and Y directions n Curve, (b) is the original two-dimensional power spectrum density PSD result, (c) is the two-dimensional power spectrum density PSD result after signal filtering and denoising, and (d) is the radial average PSD result;

[0056] Figure 4The isotropic component and the anisotropic component of the PSD domain after the component extraction operation of the embodiment of the method for extracting and identifying the surface morphology features of an optical element of the present invention, wherein (a) is the isotropic component of the PSD domain, and (b) is the anisotropic component of the PSD domain;

[0057] Figure 5 These are the surface isotropic (first column) and anisotropic (second column) morphology reconstruction results at different polishing times of an embodiment of a method for extracting and identifying surface morphology features of an optical element of the present invention, wherein (a) is the surface isotropic (first column) and anisotropic (second column) morphology reconstruction results at 0 min, (b) is the surface isotropic (first column) and anisotropic (second column) morphology reconstruction results at 20 min, (c) is the surface isotropic (first column) and anisotropic (second column) morphology reconstruction results at 60 min, (d) is the surface isotropic (first column) and anisotropic (second column) morphology reconstruction results at 100 min, and (e) is the surface isotropic (first column) and anisotropic (second column) morphology reconstruction results at 180 min;

[0058] Figure 6 The present invention is a method for extracting and identifying the surface topography of an optical element of the present invention. The surface topography detection results of a multilayer film grating and a single-point diamond turning surface are shown in FIG. 1 , wherein (a) is the surface topography detection result of a multilayer film grating, and (b) is the surface topography detection result of a single-point diamond turning surface.

[0059] Figure 7 The present invention is a method for extracting and identifying surface morphology features of a multilayer film grating, isotropic and anisotropic surface component signal extraction, wherein (a) is the average PSD result in the X direction and the Y direction, (b) is the two-dimensional surface morphology after filtering the noise, (c) is the isotropic component in the PSD domain, (d) is the anisotropic component in the PSD domain, (e) is the isotropic component reconstructed in the Cartesian domain, and (f) is the anisotropic component reconstructed in the Cartesian domain;

[0060] Figure 8 The present invention is a method embodiment for extracting and identifying surface morphology features of a single-point diamond turning surface, wherein (a) is the average PSD result in the X and Y directions, (b) is the two-dimensional surface morphology after filtering the noise, (c) is the isotropic component in the PSD domain, (d) is the anisotropic component in the PSD domain, (e) is the isotropic component reconstructed in the Cartesian domain, and (f) is the anisotropic component reconstructed in the Cartesian domain. DETAILED DESCRIPTION

[0061] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0062] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0063] As shown in the figure, a method for extracting and identifying surface morphological features of an optical element includes the following steps:

[0064] S1. Preparation of surface samples with different morphological features includes preparation of isotropic fractal surfaces using CMP, preparation of quasi-one-dimensional anisotropic surfaces using diamond turning, and preparation of two-dimensional anisotropic surfaces using ion beam polishing.

[0065] S2. Obtain two-dimensional surface height information through a surface profiler or AFM surface profile detection instrument; for example, a white light profiler (ContourGT-X3, Bruker TM ) or atomic force microscope (AFM Dimension Icon, Bruker TM ) detection to obtain the two-dimensional surface height distribution.

[0066] S3, local morphological feature recognition and filtering, including the following steps:

[0067] S31, use area bearing ratio to identify and filter random holes;

[0068] Specifically, the calculation formula of area bearing ratio is:

[0069]

[0070] Where F is the integral of the surface topography height distribution, f is the material area load ratio, and P(h≤h i ) represents the surface topography height h is less than h i The probability function of

[0071] Set a certain threshold as the standard for identifying holes, and set a higher threshold to filter the hole signal; for example, you can set the average valley depth S vk As the standard for identifying holes, -3×S vk As the threshold for filtering holes.

[0072] S32, using Hough transform to transform the scratches into MC space, identify the peaks and remove random scratches, specifically:

[0073] The surface topography is transformed from the Cartesian coordinate system to the MC space, the scratch is modeled as a straight line y=mx+c, and the scratch is identified and removed by the peak value in the MC space.

[0074] S4. Apply the Welch window to reduce sampling errors and calculate the two-dimensional power spectrum density PSD results, specifically:

[0075] The Welch window is applied to reduce the blur and leakage caused by the limited sampling length and discrete sampling points. The specific result of the two-dimensional power spectrum density PSD is:

[0076]

[0077] Among them, the reconstruction normalization function w satisfies:

[0078]

[0079] in, represents the Welch window function, Represents the rectangular window function.

[0080] S5. Use radial averaging method to obtain one-dimensional power spectral density PSD n The results, specifically:

[0081] The radial average of the two-dimensional power spectral density PSD after noise filtering is as follows:

[0082]

[0083] Among them, PSD n is the two-dimensional PSD result after noise filtering, ρ L and ρ L-1 The coordinate point (u, v) satisfies in the polar coordinate system The polar diameter, N L Indicates the number of points within the region.

[0084] S6. One-dimensional power spectral density PSD under linear fitting double logarithmic coordinates n As a result, the slope p and intercept q of the fitting curve are obtained, and the fitted PSD curve is expressed in the following form:

[0085] log(PSD n (ρ L ))=p·log(ρ L )+q.

[0086] S7, according to the one-dimensional power spectral density PSD n Results, fitting slope p and intercept q, calculation of the two-dimensional extraction factor C for the isotropic and anisotropic components iso , C anis , can be written as:

[0087]

[0088]

[0089] S8. Inverse calculation of the two-dimensional Fourier components of the isotropic and anisotropic components of the surface morphology and Specifically:

[0090] The following two-dimensional Fourier components are obtained by calculating the extraction factors in the Fourier domain:

[0091]

[0092]

[0093] S9, using inverse Fourier transform to obtain the isotropic and anisotropic two-dimensional surface morphology components h recons , the specific calculation process is:

[0094]

[0095] in, is the reconstructed two-dimensional Fourier component, and IFT stands for inverse Fourier transform; substituting the reconstructed two-dimensional Fourier component into the above formula can obtain the reconstructed surface of the corresponding component.

[0096] S10, respectively evaluating the extracted isotropic and anisotropic two-dimensional morphological components, specifically:

[0097] For the isotropic component, its fractal dimension and surface morphology roughness parameters are the main evaluation indicators; for the anisotropic component, the peak value, characteristic periodicity and directionality of the PSD curve are the main evaluation indicators.

[0098] Example 1

[0099] For Figure 2 The CMP polished fractal surface is fed as input data into the surface topography component extraction code, followed by the specific extraction process described above.

[0100] Figure 3 (a) shows Figure 2 (a) PSD curves of the surface in the X and Y directions. These two curves are quasi-linear and overlap in the Nyquist cutoff frequency region, indicating that the surface has similar fractal behavior in both directions. After noise filtering using steps S31 and S32, we obtain Figure 3 The two-dimensional power spectral density PSD shown in (c) is compared with the one before denoising. Figure 3 (b) The original two-dimensional power spectrum density PSD, the line perpendicular to the X-axis almost disappears, while the other two lines become more obvious due to the enhancement of the signal-to-noise ratio. Then, the radial average PSD is calculated according to step S5, as shown in Figure 3As shown in (d), the p and q values ​​obtained by linear fitting are -2.91 and -11.4, respectively, and R2 is 0.93, indicating that the surface is isotropic.

[0101] The component extraction operation is then performed and the extraction factors are calculated and displayed Figure 4 In. C iso Similar to the circular spots representing fractal surfaces, C anis is the information remaining after minimizing the isotropic surface component, so the circular spots are eliminated.

[0102] Figure 2 (ae) shows the surface micromorphology evolution test results when the polishing conditions are changed during the CMP process. Applying the above method to the test surface during the CMP process will be able to obtain the evolution of the isotropic and anisotropic components of the surface morphology during the polishing process, such as Figure 5 As shown in (ae), the polishing endpoint judgment and the accurate identification of the slight changes in the polishing process parameters can be achieved.

[0103] Example 2

[0104] For Figure 5 The one-dimensional grating and near-three-dimensional single-point diamond turning detection surface are fed as input data into the surface morphology component extraction code, followed by the specific extraction process described above.

[0105] Figure 7 (a) shows Figure 6 (a) PSD curves of the surface in the X and Y directions. These two curves do not overlap in the Nyquist cutoff frequency region, indicating that the surface has different structural features in the two directions. After using the noise filtering of steps S31 and S32, we get Figure 7 (b) shows the two-dimensional power spectral density PSD. Then, the radial average PSD is calculated according to step S5, and then the component extraction operation is performed, and the extraction factor is calculated and displayed Figure 7 (cd) in. iso Similar to the circular spots representing fractal surfaces, C anis is the information remaining after minimizing the isotropic surface component, so the circular spots are eliminated. Figure 7 (ef) show the reconstructed surface components, and the RMS roughness of the reconstructed isotropic and anisotropic components are 0.12nm and 4.65nm, respectively.

[0106] Figure 8 (a) shows Figure 6(b) PSD curves of the surface in the X and Y directions. These two curves do not overlap in the Nyquist cutoff frequency region, indicating that the surface has different structural characteristics in the two directions. After using the noise filtering of steps S31 and S32, we obtain Figure 8 (b) shows the two-dimensional power spectral density PSD. Then, the radial average PSD is calculated according to step S5, and then the component extraction operation is performed, and the extraction factor is calculated and displayed Figure 8 (cd). C iso Similar to the circular spots representing fractal surfaces, C anis is the information remaining after minimizing the isotropic surface component. Figure 7 (ef) show the reconstructed surface components, and the RMS roughness of the reconstructed isotropic and anisotropic components are 0.11 nm and 4.74 nm, respectively.

[0107] Therefore, the present invention adopts the above-mentioned method for extracting and identifying the surface morphology features of an optical element, which can separate isotropic and anisotropic surface components, reconstruct reliable two-dimensional morphology information with improved contrast, intuitively evaluate the morphological characteristics of each component, help improve the efficiency and accuracy of surface detection, and reduce the experience requirements for detection personnel.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for extracting and identifying surface morphology features of an optical element, characterized in that: The following steps are involved: S1. Prepare surface samples with different morphological characteristics; S2, obtaining two-dimensional surface topography height information through a surface topography detection instrument; S3, local morphological feature recognition and filtering; S4, applying the Welch window to reduce sampling errors and calculating the two-dimensional power spectrum density PSD results; S5. Use radial averaging method to obtain one-dimensional power spectral density PSD n result; S6. One-dimensional power spectral density PSD under linear fitting double logarithmic coordinates n As a result, the slope p and intercept q of the fitting curve are obtained; S7, according to the one-dimensional power spectral density PSD n Results, fitting slope p and intercept q, calculation of the two-dimensional extraction factor C of the isotropic and anisotropic components of the surface morphology iso , C anis ; S8. Inverse calculation of the two-dimensional Fourier transform components of the isotropic and anisotropic components of the surface morphology and S9, using inverse Fourier transform to obtain the isotropic and anisotropic two-dimensional surface morphology components h recons ; S10, evaluating the extracted isotropic and anisotropic two-dimensional morphological components; In step S7, the one-dimensional power spectral density PSD n Results, fitted slope and intercept, two-dimensional extraction factor C of the isotropic and anisotropic components iso , C anis writing: Among them, C iso The circular spots representing the fractal surface, C anis is the information remaining after minimizing the isotropic surface components.

2. The method for extracting and identifying surface morphology features of an optical element according to claim 1, characterized in that: In step S1, the preparation of surface samples with different morphological features includes preparing an isotropic fractal surface by using CMP, preparing a quasi-one-dimensional anisotropic surface by using diamond turning, and preparing a two-dimensional anisotropic surface by using ion beam polishing.

3. The method for extracting and identifying surface morphology features of an optical element according to claim 2, characterized in that: In step S3, the local morphological feature recognition and filtering includes the following steps: S31, use area bearing ratio to identify and filter random holes; The calculation formula of the area bearing ratio is: Where F is the integral of the surface topography height distribution, f is the material area load ratio, and P(h≤h i ) represents the surface topography height h is less than h i The probability function of S32, using Hough transform to transform the scratches into MC space, identify the peaks and remove random scratches, specifically: The surface topography is transformed from the Cartesian coordinate system to the MC space, the scratch is modeled as a straight line y=mx+c, and the scratch is identified and removed by the peak value in the MC space.

4. The method for extracting and identifying surface morphology features of an optical element according to claim 3, characterized in that: In step S4, the Welch window is applied to reduce the sampling error and calculate the two-dimensional power spectrum density PSD result, specifically: the Welch window is applied to reduce the blur and leakage caused by the limited sampling length and discrete sampling points, and the result of the calculated two-dimensional power spectrum density PSD is: Among them, the reconstruction normalization function w satisfies: in, represents the Welch window function, Represents the rectangular window function.

5. The method for extracting and identifying surface morphology features of an optical element according to claim 4, characterized in that: In step S5, the one-dimensional power spectral density PSD is obtained by using the radial averaging method. n The results, specifically: The radial average of the two-dimensional power spectral density PSD after noise filtering is as follows: Among them, PSD n is the two-dimensional PSD result after noise filtering, ρ L and ρ L-1 The coordinate point (u, v) satisfies in the polar coordinate system The polar diameter, N L Indicates the number of points within the region.

6. The method for extracting and identifying surface morphology features of an optical element according to claim 5, characterized in that: In step S6, the linear fitting one-dimensional power spectral density PSD under double logarithmic coordinates n As a result, the slope p and intercept q of the fitting curve are obtained, and the PSD after fitting is n The curve is expressed in the following form: log(PSD n (r L ))=p·log(ρ L )+q。 7. The method for extracting and identifying surface morphology features of an optical element according to claim 6, characterized in that: In step S8, the two-dimensional Fourier results of the isotropic and anisotropic components of the surface morphology are inversely calculated. and Specifically: The following two-dimensional Fourier components are obtained by calculating the extraction factors in the Fourier domain:

8. The method for extracting and identifying surface morphology features of an optical element according to claim 7, characterized in that: In step S9, the inverse Fourier transform is used to obtain the isotropic and anisotropic two-dimensional surface morphology components h recons , the specific calculation process is: in, is the reconstructed two-dimensional Fourier component, and IFT stands for inverse Fourier transform. Substituting the reconstructed two-dimensional Fourier component into the above formula, the reconstructed surface of the corresponding component is obtained.

9. The method for extracting and identifying surface morphology features of an optical element according to claim 8, characterized in that: In step S10, the extracted isotropic and anisotropic two-dimensional morphological components are evaluated respectively, specifically: For the isotropic component, its fractal dimension and surface morphology roughness parameters are the main evaluation indicators; for the anisotropic component, the peak value, characteristic periodicity and directionality of the PSD curve are the main evaluation indicators.

Images